Prospective Cohort Study of Transarterial Chemoembolization for Unresectable Hepatocellular Carcinoma in 8510 Patients

Background & Aims: To elucidate the survival of the patients with unresectable hepatocellular carcinoma (HCC) who underwent transcatheter arterial lipiodol chemoembolization (TACE) and to analyze the factors affecting the survivals. Methods: During the last 8 years, a nationwide prospective cohort study was performed in 8510 patients with unresectable HCC who underwent TACE using emulsion of lipiodol and anticancer agents followed by gelatin sponge particles as an initial treatment. Exclusion criteria were extrahepatic metastases and/or any previous treatment prior to the present TACE. The primary end point was survival. The survival rates were calculated by the Kaplan–Meier method. The multivariate analyses for the factors affecting survival were evaluated by the Cox proportional hazard model. The mean follow-up period was 1.77 years. Results: For overall survival rates by TACE, median and 1-, 3-, 5-, and 7-year survivals were 34 months, 82%, 47%, 26%, and 16%, respectively. Both the degree of liver damage and the tumor-node-metastasis (TNM) system proposed by the Liver Cancer Study Group of Japan demonstrated good stratification of survivals (P = .0001). The multivariate analyses showed significant difference in degree of liver damage (P = .0001), α-fetoprotein value (P = .0001), maximum tumor size (P = .0001), number of lesions (P = .0001), and portal vein invasion (P = .0001). The last 3 factors could be replaced by TNM stage. The TACE-related mortality rate after the initial therapy was .5%. Conclusions: TACE showed safe therapeutic modality with a 5-year survival of 26% for unresectable HCC patients. The degrees of liver damage, TNM stage, and α-fetoprotein values were independent risk factors for patient survival. Background & Aims: To elucidate the survival of the patients with unresectable hepatocellular carcinoma (HCC) who underwent transcatheter arterial lipiodol chemoembolization (TACE) and to analyze the factors affecting the survivals. Methods: During the last 8 years, a nationwide prospective cohort study was performed in 8510 patients with unresectable HCC who underwent TACE using emulsion of lipiodol and anticancer agents followed by gelatin sponge particles as an initial treatment. Exclusion criteria were extrahepatic metastases and/or any previous treatment prior to the present TACE. The primary end point was survival. The survival rates were calculated by the Kaplan–Meier method. The multivariate analyses for the factors affecting survival were evaluated by the Cox proportional hazard model. The mean follow-up period was 1.77 years. Results: For overall survival rates by TACE, median and 1-, 3-, 5-, and 7-year survivals were 34 months, 82%, 47%, 26%, and 16%, respectively. Both the degree of liver damage and the tumor-node-metastasis (TNM) system proposed by the Liver Cancer Study Group of Japan demonstrated good stratification of survivals (P = .0001). The multivariate analyses showed significant difference in degree of liver damage (P = .0001), α-fetoprotein value (P = .0001), maximum tumor size (P = .0001), number of lesions (P = .0001), and portal vein invasion (P = .0001). The last 3 factors could be replaced by TNM stage. The TACE-related mortality rate after the initial therapy was .5%. Conclusions: TACE showed safe therapeutic modality with a 5-year survival of 26% for unresectable HCC patients. The degrees of liver damage, TNM stage, and α-fetoprotein values were independent risk factors for patient survival. Hepatocellular carcinoma (HCC) is the fifth most common cancer in the world and is responsible for 500,000 deaths globally every year.1Parkin D.M. Bray F. Ferlay J. Pisani P. Estimating the world cancer burden Globocan 2000.Int J Cancer. 2001; 94: 153-156Crossref PubMed Scopus (3271) Google Scholar The prognosis still remains poor because of the advanced stage of cancer and associated hepatic impairment at diagnosis and because of the high intrahepatic recurrence rate at 5 years of 79%–80% after hepatic surgery2Imamura H. Matsuyama Y. Tanaka E. Ohkubo T. Hasegawa K. Miyagawa S. Sugawara Y. Minagawa M. Takayama T. Kawasaki S. Makuuchi M. Risk factors contributing to early and late phase intrahepatic recurrence of hepatocellular carcinoma after hepatectomy.J Hepatol. 2003; 38: 200-207Abstract Full Text Full Text PDF PubMed Scopus (1088) Google Scholar, 3Chen M.F. Hwang T.L. Jeng L.B. Wang C.S. Jan Y.Y. Chen S.C. Postoperative recurrence of hepatocellular carcinoma. Two hundred five consecutive patients who underwent hepatic resection in 15 years.Arch Surg. 1994; 129: 738-742Crossref PubMed Scopus (184) Google Scholar and of 83% after percutaneous ethanol injection (PEI),4Koda M. Murawaki Y. Mitsuda A. Ohyama K. Horie Y. Suou T. Kawasaki H. Ikawa S. Predictive factors for intrahepatic recurrence after percutaneous ethanol injection therapy for small hepatocellular carcinoma.Cancer. 2000; 88: 529-537Crossref PubMed Scopus (132) Google Scholar resulting from either intrahepatic metastases from the primary tumor or multicentric occurrence. With regard to treatment, surgical resection and PEI or radio-frequency ablation (RFA) are considered to be curable treatments of choice.5Arii S. Yamaoka Y. Futagawa S. Inoue K. Kobayashi K. Kojiro M. Makuuchi M. Nakamura Y. Okita K. Yamada R. The Liver Cancer Study Group of JapanResults of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas a retrospective and nationwide survey in Japan.Hepatology. 2000; 32: 1224-1229Crossref PubMed Scopus (695) Google Scholar Whereas transcatheter arterial embolization irrespective of with and without anticancer agent and lipiodol has been controversial in the survival benefit,6Lin D.Y. Liaw Y.F. Lee T.Y. Lai C.M. Hepatic arterial embolization in patients with unresectable hepatocellular carcinoma—a randomized controlled trial.Gastroenterology. 1988; 94: 453-456Abstract PubMed Google Scholar, 7Pelletier G. Roche A. Ink O. Anciaux M.L. Derhy S. Rougier P. Lenoir C. Attali P. Etienne J.P. A randomized trial of hepatic arterial chemoembolization in patients with unresectable hepatocellular carcinoma.J Hepatol. 1990; 11: 181-184Abstract Full Text PDF PubMed Scopus (388) Google Scholar, 8GETCHA comparison of lipiodol chemoembolization and conservative treatment for unresectable hepatocellular carcinoma.N Engl J Med. 1995; 332: 1256-1261Crossref PubMed Scopus (850) Google Scholar, 9Bruix J. Llovet J.M. Castells A. Montana X. Bru C. Ayuso M.C. Vilana R. Rodes J. Transarterial embolization versus symptomatic treatment in patients with advanced hepatocellular carcinoma results of a randomized, controlled trial in a single institution.Hepatology. 1998; 27: 1578-1583Crossref PubMed Scopus (507) Google Scholar, 10Pelletier G. Ducreux M. Gay F. Luboinski M. Hagege H. Dao T. Van Steenbergen W. Buffet C. Rougier P. Adler M. Pignon J.P. Roche A. Group CHCTreatment of unresectable hepatocellular carcinoma with lipiodol chemoembolization a multicenter randomized trial.J Hepatol. 1998; 29: 129-134Abstract Full Text PDF PubMed Scopus (388) Google Scholar2 randomized controlled trials recently elucidated that TACE showed a survival benefit compared with control.11Llovet J.M. Real M.I. Montana X. Planas R. Coll S. Aponte J. Ayuso C. Sala M. Muchart J. Sola R. Rodes J. Bruix J. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma a randomised controlled trial.Lancet. 2002; 359: 1734-1739Abstract Full Text Full Text PDF PubMed Scopus (2809) Google Scholar, 12Lo C.M. Ngan H. Tso W.K. Liu C.L. Lam C.M. Poon R.T. Fan S.T. Wong J. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma.Hepatology. 2002; 35: 1164-1171Crossref PubMed Scopus (2149) Google Scholar Transcatheter arterial embolization was initially used to treat HCC by Doyon et al13Doyon D.M.A. Jourde A.M. Regensberg C. Frileux C. L’embolisation arterielle hepatique dans les tumeurs malignes du foie.Ann Radiol. 1974; 17: 593-603PubMed Google Scholar in 1974 and was applied to most inoperable HCC using gelatin sponge particles and anticancer agents in Japan.14Yamada R. Nakatsuka H. Nakamura K. Sato M. Tamaoka K. Takemoto K. Minakuchi K. Yamaguchi S. Tamaki M. Monna T. Yamamoto S. Transcatheter arterial embolization therapy in unresectable hepatomas—experience in 15 cases.Acta Hepatologica Japonica. 1979; 20: 595-603Crossref Scopus (21) Google Scholar In the mid-1990s, lipiodol was newly introduced to enhance mainly the therapeutic effect.15Konno T. Maeda H. Iwai K. Tashiro S. Maki S. Morinaga T. Mochinaga M. Hiraoka T. Yokoyama I. Effect of arterial administration of high-molecular-weight anticancer agent SMANCS with lipid lymphographic agent on hepatoma a preliminary report.Eur J Cancer Clin Oncol. 1983; 19: 1053-1065Abstract Full Text PDF PubMed Scopus (337) Google Scholar, 16Ohishi H. Uchida H. Yoshimura H. Ohue S. Ueda J. Katsuragi M. Matsuo N. Hosogi Y. Hepatocellular carcinoma detected by iodized oil. Use of anticancer agents.Radiology. 1985; 154: 25-29PubMed Google Scholar, 17Yumoto Y. Jinno K. Tokuyama K. Araki Y. Ishimitsu T. Maeda H. Konno T. Iwamoto S. Ohnishi K. Okuda K. Hepatocellular carcinoma detected by iodized oil.Radiology. 1985; 154: 19-24PubMed Google Scholar, 18Takayasu K. Shima Y. Muramatsu Y. Moriyama N. Yamada T. Makuuchi M. Hasegawa H. Hirohashi S. Hepatocellular carcinoma treatment with intra-arterial iodized oil with and without chemotherapeutic agents.Radiology. 1987; 163: 345-351PubMed Google Scholar Transcatheter arterial lipiodol chemoembolization (TACE) is now the mainstay of treatment of choice for noncurative HCC.19Sasaki Y. Imaoka S. Kasugai H. Fujita M. Kawamoto S. Ishiguro S. Kojima J. Ishikawa O. Ohigashi H. Furukawa H. et al.A new approach to chemoembolization therapy for hepatoma using ethiodized oil, cisplatin, and gelatin sponge.Cancer. 1987; 60: 1194-1203Crossref PubMed Scopus (228) Google Scholar, 20Kasugai H. Kojima J. Tatsuta M. Okuda S. Sasaki Y. Imaoka S. Fujita M. Ishiguro S. Treatment of hepatocellular carcinoma by transcatheter arterial embolization combined with intraarterial infusion of a mixture of cisplatin and ethiodized oil.Gastroenterology. 1989; 97: 965-971PubMed Google Scholar, 21Nakamura H. Hashimoto T. Oi H. Sawada S. Transcatheter oily chemoembolization of hepatocellular carcinoma.Radiology. 1989; 170: 783-786PubMed Google Scholar, 22Uchida H. Ohishi H. Matsuo N. Nishimine K. Ohue S. Nishimura Y. Maeda M. Yoshioka T. Transcatheter hepatic segmental arterial embolization using lipiodol mixed with an anticancer drug and Gelfoam particles for hepatocellular carcinoma.Cardiovasc Intervent Radiol. 1990; 13: 140-145Crossref PubMed Scopus (196) Google Scholar, 23Matsui O. Kadoya M. Yoshikawa J. Gabata T. Arai K. Demachi H. Miyayama S. Takashima T. Unoura M. Kogayashi K. Small hepatocellular carcinoma treatment with subsegmental transcatheter arterial embolization.Radiology. 1993; 188: 79-83PubMed Google Scholar, 24Kawai S. Tani M. Okamura J. Ogawa M. Ohashi Y. Monden M. Hayashi S. Inoue J. Kawarada Y. Kusano M. Kubo Y. Kuroda C. Sakata Y. Shimamura Y. Jinno K. Takahashi A. Takayasu K. Tamura K. Nagasue N. Nakanishi Y. Makino M. Masuzawa M. Yumoto Y. Mori T. Oda T. The Cooperative Study Group for Liver Cancer Treatment of JapanProspective and randomized clinical trial for the treatment of hepatocellular carcinoma—a comparison between L-TAE with farmorubicin and L-TAE with adriamycin preliminary results (second cooperative study).Cancer Chemother Pharmacol. 1994; 33: S97-S102PubMed Google Scholar, 25Chang J.M. Tzeng W.S. Pan H.B. Yang C.F. Lai K.H. Transcatheter arterial embolization with or without cisplatin treatment of hepatocellular carcinoma. A randomized controlled study.Cancer. 1994; 74: 2449-2453Crossref PubMed Scopus (137) Google Scholar, 26Bronowicki J.P. Vetter D. Dumas F. Boudjema K. Bader R. Weiss A.M. Wenger J.J. Boissel P. Bigard M.A. Doffoel M. Transcatheter oily chemoembolization for hepatocellular carcinoma. A 4-year study of 127 French patients.Cancer. 1994; 74: 16-24Crossref PubMed Scopus (183) Google Scholar, 27Stefanini G.F. Amorati P. Biselli M. Mucci F. Celi A. Arienti V. Roversi R. Rossi C. Re G. Gasbarrini G. Efficacy of transarterial targeted treatments on survival of patients with hepatocellular carcinoma. An Italian experience.Cancer. 1995; 75: 2427-2434Crossref PubMed Scopus (121) Google Scholar, 28Hatanaka Y. Yamashita Y. Takahashi M. Koga Y. Saito R. Nakashima K. Urata J. Miyao M. Unresectable hepatocellular carcinoma analysis of prognostic factors in transcatheter management.Radiology. 1995; 195: 747-752Crossref PubMed Scopus (103) Google Scholar, 29Chung J.W. Park J.H. Han J.K. Choi B.I. Han M.C. Lee H.S. Kim C.Y. Hepatic tumors predisposing factors for complications of transcatheter oily chemoembolization.Radiology. 1996; 198: 33-40PubMed Google Scholar, 30Solomon B. Soulen M.C. Baum R.A. Haskal Z.J. Shlansky-Goldberg R.D. Cope C. Chemoembolization of hepatocellular carcinoma with cisplatin, doxorubicin, mitomycin-C, ethiodol, and polyvinyl alcohol prospective evaluation of response and survival in a US population.J Vasc Interv Radiol. 1999; 10: 793-798Abstract Full Text PDF PubMed Scopus (97) Google Scholar, 31Savastano S. Miotto D. Casarrubea G. Teso S. Chiesura-Corona M. Feltrin G.P. Transcatheter arterial chemoembolization for hepatocellular carcinoma in patients with Child’s grade A or B cirrhosis a multivariate analysis of prognostic factors.J Clin Gastroenterol. 1999; 28: 334-340Crossref PubMed Scopus (75) Google Scholar, 32Takayasu K. Muramatsu Y. Maeda T. Iwata R. Furukawa H. Moriyama N. Okusaka T. Okada S. Ueno H. Targeted transarterial oily chemoembolization for small foci of hepatocellular carcinoma using a unified helical CT and angiography system analysis of factors affecting local recurrence and survival rates.AJR Am J Roentgenol. 2001; 176: 681-688Crossref PubMed Scopus (140) Google Scholar The latest nationwide report registered every 2 years by the Liver Cancer Study Group of Japan33Ikai I. Arii S. Ichida T. Okita K. Omata M. Kojiro M. Takayasu K. Nakanuma Y. Makuuchi M. Matsuyama Y. Yamaoka Y. Report of the 16th follow-up survey of primary liver cancer.Hepatol Res. 2005; 32: 163-172Crossref PubMed Scopus (107) Google Scholar showed the frequency of treatment for HCC as an initial therapy: transcatheter arterial embolization including TACE and transcatheter arterial injection of emulsion of lipiodol and anticancer agent (without gelatin sponge) accounted for 36.4% of 16,941 patients; surgical resection for 31.3%; local ablation therapy such as PEI, microwave coagulation therapy, and RFA for 26.8%; and others for 5.5%. Moreover, TACE serves for treatment of intrahepatic recurrent foci in 52% of patients who developed after previous treatments.33Ikai I. Arii S. Ichida T. Okita K. Omata M. Kojiro M. Takayasu K. Nakanuma Y. Makuuchi M. Matsuyama Y. Yamaoka Y. Report of the 16th follow-up survey of primary liver cancer.Hepatol Res. 2005; 32: 163-172Crossref PubMed Scopus (107) Google Scholar To date, the long-term survival rate and analysis of predictor affecting the survival rate of TACE for more than 8000 patients with HCC have not yet been reported. Thus, we conducted a study to elucidate them using a list of patients registered nationwide. During the last 8 years from January 1994 to December 2001, a total of 72,866 patients with primary liver cancer were newly registered biannually by the Liver Cancer Study Group of Japan throughout some 800 medical institutions using a registration/questionnaire sheet with more than 180 questions. HCC patients occupied 95% (69,182/72,866) of primary liver cancer. Of these, 8542 patients were assigned in the current prospective cohort study, who underwent TACE as an initial treatment for unresectable HCC and did not receive any other therapy during the first investigation period of no more than 2 years and fulfilled the following inclusion criteria. First, the hepatic lesion was clinicopathologically diagnosed as HCC. Other primary hepatic lesions such as intrahepatic cholangiocarcinoma, sarcoma, and other benign lesions were excluded. Second, there was neither extrahepatic metastasis to lymph nodes and/or other organs nor any treatment prior to the present TACE. Finally, 8510 patients were enrolled, with the exclusion of 32 patients whose replies for the questionnaire were incomplete. The diagnosis of HCC was made mainly based on the reliable imaging modalities using ultrasonography, dynamic computed tomography (CT), magnetic resonance imaging (MRI), and angiography and/or pathologically by biopsy specimens (4.5%). Abnormal elevation of tumor markers was also referred: α-fetoprotein more than 400 ng/mL (normal, <20 ng/mL) and des-γ-carboxyl prothrombin more than 100 mAU/mL (normal, <40 mAU/mL). Typical HCC was depicted as hyperattenuation in arterial phase and hypoattenuation or washout in delayed phase (approximately 3 minutes after the contrast injection) of dynamic CT34Araki T. Itai Y. Furui S. Tasaka A. Dynamic CT densitometry of hepatic tumors.AJR Am J Roentgenol. 1980; 135: 1037-1043Crossref PubMed Scopus (127) Google Scholar and on dynamic MRI, as was shown a hypervascular lesion on hepatic arteriography. The extrahepatic metastases were routinely examined by ultrasonography, CT, and chest x-ray. The distribution of background factors was tabulated in Table 1. The study population consisted of 6122 males and 2386 females with a mean age of 66.2 years (standard deviation (SD), 9.15). The number of patients in grades A, B, and C of degree of liver damage was 4008 (51%), 3053 (39%), and 766 (10%), respectively. The degree of liver damage proposed by the Liver Cancer Study Group of Japan was classified as grades A, B, and C based on the highest grade containing at least 2 findings such as ascites, serum bilirubin, serum albumin, indocyanine green retention rate at 15 minutes (ICG R15), and prothrombin activity (Table 2).35The Liver Cancer Study Group of JapanThe general rules for the clinical and pathological study of primary liver cancer. Version 4. Kanehara & Co., Ltd, Tokyo2000Google Scholar, 36The Liver Cancer Study Group of JapanThe general rules for the clinical and pathological study of primary liver cancer. Version 2. Kanehara & Co., Ltd, Tokyo2003Google Scholar This classification is closely related to the Child–Pugh class.Table 1Distribution of Background Factors and Results of the Univariate Analysis in 8510 Patients With Unresectable Hepatocellular Carcinoma Who Underwent Transcatheter Arterial Lipiodol ChemoembolizationBackground factorsNumber of patientsProportion (%)Survival (%)P value1 y3 y5 yAge, y <60187122784429.07 ≥60663978834725Sex Male612272814625.002 Female238628845028Degree of liver damage A400851875633 B305339804121.0001 C7661063238HBV and HCV HCV Ab positive606374844825.0001 HBs Ag positive90011733924 Both positive2113805032 Both negative97112784528Maximum tumor size (cm) ≤2198924936339.0001 2.1–3198124905228 3.1–5231828834323 ≥5.1207625633016No. of lesions 1364844875733 2–3267532844524.0001 43.0Serum albumin (g/dL)>3.53.0–3.5<3.0ICGR15 (%)<1515–40>40Prothrombin activity (%)>8050–80<50NOTE. If 2 or more items scoring the same grade occurred in the 2 grades, the higher grade is adopted as the degree of liver damage.ICGR15 (%), indocyanine green retention rate at 15 minutes. Open table in a new tab Vp, vascular invasion of the portal vein; Vv, vascular invasion of the hepatic vein. NOTE. If 2 or more items scoring the same grade occurred in the 2 grades, the higher grade is adopted as the degree of liver damage. ICGR15 (%), indocyanine green retention rate at 15 minutes. Seventy-four percent of patients were positive for hepatitis C virus antibody, 11% were positive for hepatitis B virus surface antigen, 3% were positive for both, and 12% were negative for both. Regarding the maximum tumor size, 24% of patients each had lesion(s) no larger than 2 cm in diameter and lesion(s) measured from 2.1 to 3 cm, 28% of patients had lesion(s) from 3.1 to 5 cm, and 25% had lesion(s) larger than 5.1 cm. The number of lesions was 1 in 44% of patients, 2 or 3 in 32% of patients, and more than 4 in 25% of patients. The degree of vascular invasion to the portal vein was Vp0 (no invasion) in 88% of patients, and in the remaining 12% of patients, Vp1 (invasion to the third order or more peripheral branch) in 4%, Vp2 (the second order portal vein) in 4%, and more than Vp3 (the first portal vein or main portal trunk) in 4% of patients. Meanwhile, the degree of hepatic vein invasion was Vv0 (no invasion) in 97% of patients and more than Vv1 (any hepatic vein invasion including the main hepatic veins and/or inferior vena cava) in 3% of patients. Normal AFP value was seen in 34% of patients, 21–200 ng/mL in 35%, 201–400 ng/mL in 7%, and more than 401 ng/mL in 25%. The des-γ-carboxyl prothrombin value of more than 101 mAU/mL was recognized in 46% of patients. The tumor-node-metastasis (TNM) staging adopted in the present study was revised by the Liver Cancer Study Group of Japan in 2000 (Table 3)35The Liver Cancer Study Group of JapanThe general rules for the clinical and pathological study of primary liver cancer. Version 4. Kanehara & Co., Ltd, Tokyo2000Google Scholar, 36The Liver Cancer Study Group of JapanThe general rules for the clinical and pathological study of primary liver cancer. Version 2. Kanehara & Co., Ltd, Tokyo2003Google Scholar and was proposed as a new concordant TNM classification of the primary liver cancer by the International Hepato-Pancreato-Biliary Association.37Makuuchi M. Belghiti J. Belli G. Fan S.T. Lau J.W. Ringe B. Strasberg S.M. Vauthey J.N. Yamaoka Y. Yamasaki S. IHPBA concordant classification of primary liver cancer working group report.J Hepatobiliary Pancreat Surg. 2003; 10: 26-30PubMed Google Scholar Namely, T category consists of 3 factors such as a single lesion, no larger than 2 cm in diameter, and no vascular or bile duct invasion. T stage is determined based on the factors that are fulfilled from zero to 3. Each stage of I, II, III, and IV-A corresponds to the T factor. Stage IV-B is excluded because of the inclusion criterion. The present study showed 13% of patients in stage I, 40% each in stages II and III, and 7% in stage IV-A.Table 3Definitions of TNM Stage Proposed by the Liver Cancer Study Group of JapanT factor T1Fulfilling 3 factors T2Fulfilling 2 factors T3Fulfilling 1 factor T4Fulfilling 0 factorStages Stage IT1 N0 M0 Stage IIT2 N0 M0 Stage IIIT3 N0 M0 Stage IV-AT4 N0 M0, or any T N1 M0 Stage IV-BAny T N0-1 M1T factor consists of I, Single; II, <2 cm; III, no vascular involvement Open table in a new tab T factor consists of I, Single; II, <2 cm; III, no vascular involvement The catheter tip was advanced at the nearest site of the feeding artery as possible. The emulsion of anticancer agent and lipiodol followed by gelatin sponge particles was carefully injected in the x-ray monitoring. The dose of emulsion of anticancer agent and lipiodol and the pieces of embolic materials used for TACE were determined based on the tumor size and extension of the lesions. The patients were followed by dynamic CT or MRI every 3 to 4 months, and repeated TACE was determined when the local recurrence, intrahepatic metastases, and/or second primary HCC was found. The survival rate of patients who underwent TACE was calculated from the date of diagnosis of HCC and the follow-up was ended on December 31, 2001. All patient deaths were the end point irrespective of cause of death. The mean follow-up period was 1.77 years (range, 0.003–7.99). TACE-related death was designated as death within 30 days of the initial therapy. The 11 variables listed in Table 1 were analyzed via univariate analysis. The multivariate analysis was performed by the Cox proportional hazard model. The survival rates were obtained by the Kaplan–Meier method and compared by the log-rank test. All variables with a P value less than .05 by univariate analysis were subjected to multivariate analysis. All significance tests were 2-tailed, and a P value less than .05 was considered statistically significant. All statistical analyses were performed with the Statistical Analysis System (SAS) version 8.02 (SAS Inc, Cary, NC). During an 8-year follow-up period, the total numbers of patient deaths and censored cases were 3605 and 4905, respectively. The number of patients who died and were censored annually was as follows: 1316 and 1939 during year 1, 1067 and 1258 from 1 to 2 years, 630 and 764 from 2 to 3 years, 341 and 454 from 3 to 4 years, 148 and 216 from 4 to 5 years, 72 and 145 from 5 to 6 years, 23 and 81 from 6 to 7 years, and 8 and 48 from 7 to 8 years, respectively. For overall survival of the 8510 patients who underwent TACE, the median and 1-, 3-, 5-, and 7-year survival rates were 34 months, 82%, 47%, 26%, and 16%, respectively (Table 4) (Figure 1). With the degree of liver damage classification, 5-year survival rates of grades A, B, and C were 33%, 21%, and 8%, respectively, with statistical significance (P = .0001) (Figure 2). According to the TNM staging system, 5-year survival rates in stages I, II, III, and IV-A were 47%, 32%, 20%, and 10%, respectively, with significant difference among them (P = .0001) (Figure 3).Table 4The Survival of Overall Patients Who Underwent Transcatheter Arterial Lipiodol Chemoembolization, Based on Degree of Liver Damage, TNM Stage, and Combination of Degree of Liver Damage and TNM StageGrading/stagingnSurvival (%)Median (mo)P value1 y2 y3 y4 y5 y7 yOverall survival851082634734261634Degree of liver damage (n = 7827) A400887715642332141 B305380594129211230.0001 C766633723158—17TNM stage by Liver Cancer Study Group of Japan (n = 7311) I92796867257473056.0001 II293490735743322242 III294978563926201129 IV-A5014927161010—12Combination of degree of liver damage and TNM stageLiver damage A (n = 3499) I48998927864523862.0001 II143994806652392650 III135884644733271435 IV-A2135934241210—15Liver damage B (n = 2667).0001 I30992826750431647 II106888705238281638 III11167954322215826 IV-A1744521131010—11Liver damage C (n = 648).0001 I599469523724—39 II2247249291911—24 III282663023149—17 IV-A8327163———6 Open table in a new tab Figure 2Comparison of survival rates stratified by degrees of liver damage A, B, and C with statistical significance (P = .0001).View Large Image Figure ViewerDownload (PPT)Figure 3Comparison of survival rates stratified by TNM stages I, II, III, and IV-A with statistical significance (P = .0001).View Large Image Figure ViewerDownload (PPT) Next, the survival rates of patients were estimated using a combination of degree of liver damage and the TNM staging system. In the degree of liver damage grade A, 5-year survival rates in stages I, II, III, and IV-A were 52%, 39%, 27%, and 10%, respectively, with significant differences among 4 subgroups (P = .0001). The well-stratified survival curves were obtained (not shown). The same tendency as grade A was recognized in liver damage grades B and C (P = .0001). On the other hand, 5-year survival rates in TNM stage I through liver damage grades A, B, and C were 52%, 43%, and 24%, respectively, with significant differences (P = .0001). The same results were recognized in stages II, III, and IV-A through the liver damage grades A, B, and C (P = .0001). The univariate analysis revealed the following 11 factors as independent prognostic variables: gender, degree of liver damage, hepatitis B virus surface antigen, hepatitis C virus antibody, maximum tumor size, number of lesions, portal vein invasion, hepatic vein invasion, α-fetoprotein (AFP) value, des-γ carboxyl prothrombin value, and TNM stage (Table 1). The multivariate analysis showed the following 5 factors as an independent prognostic factor: degree of liver damage, maximum tumor size, number of lesion(s), degree of portal vein invasion, and AFP value in trial 1 (Table 5). The hazard ratio of each variable ranged from 1.48 to 2.52, among which the degree of liver damage between grades A and C followed by portal vein invasion had the highest hazard ratio of 2.52 and 2.27, respectively.Table 5Multivariate Analysis of Factors Affecting Survival of Patients Who Underwent Transcatheter Arterial Lipiodol Chemoembolization Using Cox Proportional Hazard Model in Trials 1 and 2VariablesTrial 1Trial 2P valueHazard ratio (95% CI)P valueHazard ratio (95% CI)Age, y, ≥60 vs <60.421.04 (.95–1.14).361.05 (.95–1.15)Sex, Male vs female.161.07 (.98–1.16).191.06 (.97–1.16)Degree of liver damage B vs A.00011.49 (1.37–1.61